Distinct Effects of Tissue-Type Plasminogen Activator and SMTP-7 on Cerebrovascular Inflammation Following Thrombolytic Reperfusion
Background and Purpose—Thrombolysis therapy using tissue-type plasminogen activator (t-PA) is occasionally accompanied by harmful outcomes, including intracerebral hemorrhage. We have reported that Stachybotrys microspora triprenyl phenol-7 (SMTP-7), a candidate thrombolytic drug, has excellent therapeutic effect on cerebral infarction in embolic stroke model in mice; however, little is known regarding whether this agent influences cerebrovascular inflammation following thrombolytic reperfusion. The current study aimed to compare the effects of recombinant t-PA (rt-PA) and SMTP-7 on cerebrovascular inflammation.
Methods—The impact of rt-PA- and SMTP-7-induced thrombolytic reperfusion on leukocyte dynamics was investigated in a photochemically induced thrombotic middle cerebral artery occlusion (tMCAo) model in mice.
Results—Both rt-PA and SMTP-7 administration in tMCAo mice (each 10 mg/kg) resulted in thrombolytic reperfusion. The SMTP-7-administered mice showed relatively mild rolling and attachment of leukocytes to the vascular wall in the middle cerebral vein, with weak peroxynitrite reactions and proinflammatory gene expression (IL-1β, TNF-α, ICAM-1, and VCAM-1); thus, a small infarct volume compared with rt-PA-administered mice. In vitro study suggested that rt-PA at 20 μg/mL, but not SMTP-7 at a similar concentration, promotes cytokine-induced reactive oxygen species generation in cultured endothelial cells; moreover, SMTP-7 suppressed cytokine-induced VCAM-1 induction in the cells and leukocyte/ endothelial cell adhesions.
Conclusions—Relatively mild cerebrovascular inflammation and cerebral infarction in the SMTP-7 mice, compared with in rt-PA mice, is thought to be caused at least in part by direct antioxidative actions of SMTP-7 in ECs.
Thrombolysis therapy using tissue plasminogen activator (t-PA) is known to be greatly effective for acute ischemic stroke in some instances. It has been, however, reported that the t-PA therapy is occasionally accompanied by harmful outcomes, including intracerebral hemorrhage.1 A growing body of evidence suggests that the adverse effects of t-PA are at least in part caused by t-PA-induced direct cytotoxic responses in the ischemic core and surrounding penumbra, as well as ischemia/reperfusion injury, which is indirectly induced by thrombolytic reperfusion.2
The cytotoxic effects of t-PA have been studied both in animal models and on the cellular level. Wang et al demonstrated that deficiency of the t-PA gene in mice results in reduction of middle cerebral artery occlusion (MCAo)-induced cerebral infarction.3 Furthermore, exogenous administration of recombinant t-PA (rt-PA) reportedly enhances MCAo-induced cerebral infarction in mice, as well as hypoxia/reoxygenation (H/R)- and NMDA-induced apoptosis in endothelial cells (EC) and neuronal cells, respectively; this suggests that t-PA is capable of enhancing cytotoxic responses in cells.4
Brain ischemia and later reperfusion elicit an inflammatory response, in particular in brain microcirculation. ECs, astrocytes, microglial cells, and leukocytes are reportedly activated in response to ischemia/reperfusion, and release proinflammatory cytokines in a redox-sensitive manner.5 These cytokines play a key role in the induction of cell adhesion molecules, which contribute to leukocyte adhesion to the vascular wall and infiltration.6 Several studies have shown that cerebral infarction induced by transient MCAo is significantly reduced when leukocyte adhesion to ECs is inhibited7,8; thus, this process is implicated largely in secondary neuronal destruction. To our knowledge, direct data regarding t-PA-induced cerebrovascular inflammation in animal models has not been yet reported; however, it is expected that t-PA worsens this process, as t-PA reportedly facilitates inflammatory responses through activation of NF-κB in ECs.4
We have previously reported isolation of the triprenyl phenol staplabin analog Stachybotrys microspora triprenyl phenol-7 (SMTP-7), a low-molecular-weight fibrinolytic agent from cultures of Stachybotrys microspora IFO30018.9 This compound, which includes a vitamin E (Ve)-like structure, promotes urokinase-catalyzed conversion of plasminogen to plasmin, fibrin binding to plasminogen, and fibrinolysis9; this results in amelioration of embolic stroke in Mongolian gerbil and mouse models.10,11 As functional mechanisms of this agent have not been fully elucidated, the aim of this study is to compare the impact of t-PA- with SMTP-7-induced thrombolysis on cerebrovascular inflammation in the ischemic core in a photochemically induced thrombotic MCAo (tMCAo) model in mice. These data showed relatively mild cerebrovascular inflammation and cerebral infarction in SMTP-7-administered tMCAo mice compared with rt-PA-administered tMCAo mice. Our in vitro data suggested that SMTP-7 directly attenuates leukocyte/EC interaction by inhibiting cytokine-induced oxidative responses in ECs.
Materials and Methods
Photothrombosis and Thromboysis in tMCAo Mice
The animal protocols were approved by the Institutional Animal Care Committee. Male ddY mice (25 g to 35 g) were anesthetized with isoflurane inhalation (1.5% to 1.8%), and a 1 cm incision was made vertically midway between the right orbit and the right external auditory canal. The temporalis muscle was separated and retracted, and a skull hole 5 mm in diameter was made under a stereomicroscope, resulting in exposing the distal segment of the middle cerebral artery (MCA) and superficial middle cerebral vein (sMCV). Then, the animal was mounted onto the microscopic stage of a confocal microscope (LSM 510, Carl Zeiss) and the exposed cortex was held on a coverslip. To induce photothrombosis, we utilized a diode laser (wavelength: 561 nm; 15 mW of output power) and a laser-scanning unit equipped with confocal microscopy, as previously described by Ding et al.12 Following intravenous administration of the photosensitizer Rose Bengal (40 mg/kg), the arterial branch of interest was irradiated by the diode laser for 10 minutes using a laser-scan unit equipped with microscopy. For administration of rt-PA (10 mg/kg in total) or SMTP-7 (10 mg/kg in total), 10% of the thrombolytic agents were intravenously administered to the mice 30 minutes following photothrombosis, followed by an intravenous infusion of the remaining agents for 30 minutes.
2,3,5-Triphenyltetrazolium Chloride Staining and Immunohistochemistry
Twenty-four hours following photothrombosis, 2,3,5-Triphenyltetrazolium chloride staining was conducted as described previously.11 For immunohistochemical analysis, presectioned brain slices (2 mm thick) were cryosectioned at 20-μm thick. Following endogenous horseradish peroxidase inactivation and blocking by 2.5% bovine serum albumin, sections were incubated in an antibody raised against anti-4-hydroxy-2-nonenal (4-HNE)-labeled proteins (Alpha Diagnostic International) or in a nonimmune rabbit antibody (Santa Cruz). Bound antibody was detected by using the EnVision kit (Dako) followed by counterstaining with hematoxylin.
Observation of Leukocyte Dynamics
To label fluorescently ECs and leukocytes, 0.1 mL of 0.1% (wt/vol) acridine orange in isotonic saline was injected into the tail vein. MCV images in the cortex surface were obtained by using confocal microscopy (LSM 510, Carl Zeiss) equipped with ×20 objective lens (NA0.75) at 2.3-ms intervals. Leukocyte velocity, number of rolling leukocytes, and adherent leukocytes per unit-vessel-area were quantified as an index of cerebrovascular inflammation. At least 3 vessel regions per animal were measured.
Real-Time Reverse Transcription Polymerase Chain Reaction (Real-Time RT-PCR)
Ischemic core approximately 5 mm in diameter was excised during the time period following reperfusion. Complementary DNA, which was converted from total RNA extracted from excised brain segments, served as a template for qPCR reaction. The nucleotide sequence of primers is shown in supplemental Table 1 (http://stroke.ahajournals.org).
Confluent human umbilical vein endothelial cells (HUVECs) from 2 to 7 passages (Cambrex) were utilized in the experiments as previously described.13 The cells were treated with thrombolytic agents or Ve in the presence or absence of proinflammatory cytokines TNF-α or IL-1β for 3 hours. To subject HUVECs to hypoxic conditions, the culture medium was replaced by 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES)-buffered Dulbecco's modified minimum essential medium (pH 7.4) bubbled with N2 gas. The HUVEC culture was then bubbled with N2 gas for 30 minutes, and was subsequently replaced with the Dulbecco's modified minimum essential medium containing O2. Following 3 hours of a reoxygenation period, the cells were subjected to RT-PCR analysis. Human monocytic cell line THP-1 cells (DS Pharma Biomedical) were cultured in Rosewell Park Memorial Institute medium 1640 supplemented with 10% fetal calf serum.
THP-1 cells were labeled with calcein red-orange-acetoxymethyl ester (AM) at 1 μmol/L for 10 minutes at 37°C. Following stimulation, ECs were labeled with calcein/AM at 1 μmol/L for 10 minutes at 37°C, followed by coincubation with THP-1 cells (5×105 cells/well) for 30 minutes at 37°C. Nonadhering THP-1 cells were then removed, and the cells were washed twice with Krebs-HEPES buffer. Subsequently, THP-1 cells adhered to the EC monolayer were imaged by confocal microscopy (LSM510; Carl Zeiss). The mean density of adherent cells was determined by counting the THP-1 cells in 3 different fields (1.8 mm2).
Measurement of Reactive Oxygen Species
Intracellular reactive oxygen species (ROS) generation was measured using 2′,7′-dichlorofluorescein diacetate, a redox-sensitive fluorescent dye. Following stimulation, ECs were loaded with 2′,7′-dichlorofluorescein diacetate (5 μmol/L) for 30 minutes at 37°C, and then, were imaged by confocal microscopy (LSM510; Carl Zeiss). Fluorescent images of 3 different fields (0.81 mm2), which were randomly selected per well, were obtained, and mean fluorescent intensity of the images was served as a score of each well.
Immunoblotting was conducted as previously described.14 Primary antibodies used in immunoblotting are displayed in supplemental Table 2.
The nucleotide sequence of primers was shown in supplemental Table 1.
Results are expressed as mean±SEM. The Mann–Whitney U test was employed to determine the statistical difference between 2 groups. Multiple comparisons were conducted with Dunnett's test. P<0.05 was considered statistically significant in this study.
Effects of Thrombolytic Agents on tMCAo Mice
A photo-sensitizer, Rose Bengal, was administered to mice, and subsequently a green laser was precisely irradiated to the branch in distal MCAs, utilizing a laser-scanning device (Figure 1). Following 10 minutes of irradiation, clots were developed in the vessel branches (Figure 1A). We validated that blood flow in the arteries downstream of the thrombolyzed branch in all animals used in this study was stopped by the clots. After the 30-minute ischemia period, tMCAo mice were given either rt-PA at 10 mg/kg or SMTP-7 at 10 mg/kg. Our previous report indicated that cerebral infarction in the murine embolic stroke model was abolished by SMTP-7 administration at 10 mg/kg11; thus, we chose this dose to exert maximal thrombolytic effects in current animal experiments. Quantification of clot size indicated that thrombolysis in the rt-PA- or SMTP-7-administered mice is of similar degree (Figure 1B). Thrombolytic reperfusion in MCAs was observed in 6 of the 8 rt-PA mice examined, and in all of the SMTP-7 mice examined (Figure 1C; supplemental Table 3). Time required for initial reperfusion and blood velocity in MCAs in pre- and post-thrombolytic states were identical between the rt-PA and SMTP-7 mice.
The effects of the thrombolytic agents on tMCAo-induced cerebral infarction were evaluated (Figure 2A). Accordingly, rt-PA administration to tMCAo mice did not ameliorate, but rather accelerated cerebral infarction following reperfusion compared with saline-administered mice. In contrast, SMTP-7 administration did not promote cerebral infarction, but instead recovered MCA flow in a similar manner. Behavioral neurological defects were not observed in these animals (supplemental Table 4), probably because of small infarct volumes in our animal model. Furthermore, 4-HNE adducts in the ischemic core were immunohistochemically detected to evaluate peroxynitrite-mediated free radical reactions (Figure 2B). As a result, marked 4-HNE immunoreactivity, by comparison with contralateral brain cortex, was observed in the rt-PA-administered mice, but not in the SMTP-7-administered mice.
The effects of thrombolytic agents on leukocyte/EC dynamics in sMCV were investigated (Figure 3A and B). Intravital microscopy analysis indicated that tMCAo does not influence leukocyte velocity and leukocyte/EC adhesion, at least during the 3 hours following photothrombosis. Administration of rt-PA in tMCAo mice at 10 mg/kg resulted in decreasing leukocyte velocity and increasing leukocyte adhesion in sMCVs within 3 hours following rt-PA administration; in contrast, SMTP-7 administration in the mice at a similar dose did not significantly facilitate leukocyte/EC interactions.
Subsequently, inhibitory effects of thrombolytic agents on proinflammatory genes in the ischemic core were analyzed (Figure 3C). QPCR analysis indicated that tMCAo upregulates TNF-α, IL-6, IL-1β, ICAM-1, and E-selectin expression. Administration of rt-PA in tMCAo mice led to induction of VCAM-1 expression as well as further accumulation of TNF-α, IL-1β, and ICAM-1 expression; conversely, SMTP-7 administration resulted in reduction of the tMCAo-induced proinflammatory genes. Both SMTP-7 and rt-PA administration neutralize tMCAo-induced IL-6 and E-selectin overexpression.
Direct Effects of Thrombolytic Agents on Inflammatory Responses in ECs
Direct effects of thrombolytic agents on leukocyte/EC interactions were investigated using a THP-1 cell/HUVEC adhesion assay (Figure 4). Treatment of the IL-1β- or TNF-α-stimulated ECs with SMTP-7 at 20 μg/mL, but not with rt-PA at a similar concentration, led to decreasing leukocyte adhesion to ECs (Figure 4A). The SMTP-7-induced inhibition was concentration-dependent, with an IC50 of 6.1 μg/mL (Figure 4B). In addition, Ve at 20 μg/mL did not elicit further reduction in the maximal inhibition of SMTP-7, whereas administration of Ve alone at a similar concentration displayed inhibitory effects.
The impact of thrombolytic agents on endothelial adhesion molecules was evaluated in cultured ECs (Figure 5). RT-PCR analysis suggested that IL-1β and TNF-α administration as well as H/R stimulus resulted in an upregulation of the adhesion molecules ICAM-1 and VCAM-1 in the cells. Treatment of IL-1β-stimulated cells with SMTP-7 at a concentration of 20 μg/mL selectively suppressed VCAM-1 induction; in contrast, rt-PA at a similar concentration did not influence cytokine responses. This was further supported by the results derived from immunoblot analysis. Conversely, SMTP-7 treatment influenced to a lesser extent VCAM-1 overexpression in TNF-α- or H/R-stimulated cells compared with in the IL-1β-stimulated cells. Treatment of the cells with rt-PA accelerated the H/R-induced VCAM-1 expression.
The antioxidative effects of thrombolytic agents were assessed in cultured ECs (Figure 6). DCF analysis showed that both IL-1β and TNF-α stimulus facilitated ROS generation in the cells. Treatment of IL-1β-stimulated cells with rt-PA did not influence ROS generation; in contrast, SMTP-7 administration abolished IL-1β-induced ROS generation. In contrast, although the treatment of TNF-α-stimulated cells with rt-PA accelerated ROS generation, SMTP-7 administration partly inhibited TNF-α-induced ROS generation. Treatment of the cells with rt-PA or SMTP-7 alone did not influence ROS generation in the cells.
In general, cerebral ischemia/reperfusion is known to exert cytotoxic effects on the component cells, including neuronal, glial, and vascular cells. SMTP-7 reportedly facilitates the fibrinolytic system9; thus, we expected that ischemia/reperfusion-induced cerebrovascular inflammation would be observed in the rt-PA- and SMTP-7-administered tMCAo mice. However, this study surprisingly indicated that SMTP-7 does not facilitate peroxynitrite reactions, leukocyte/EC interaction, or proinflammatory gene induction, and thereby there is a lack of acceleration of tMCAo-induced cerebral infarction (Figure 2, 3). In contrast, acceleration of proinflammatory reactions and cerebral infarction was detected in the rt-PA-administered tMCAo mice (Figure 2, 3). Current in vitro data further suggested that SMTP-7 directly antagonizes IL-1β-induced VCAM-1 expression (Figure 5), ROS generation in ECs (Figure 6), and leukocyte/EC interaction (Figure 4). Furthermore, maximal inhibition of SMTP-7 on cytokine-induced leukocyte/EC adhesion is not influenced by Ve at a relatively higher concentration (Figure 4). Theriault et al documented that anti-inflammatory effects of Ve in ECs is achieved through inhibition of NF-κB activation in a redox-sensitive manner.15 In addition, administration of Ve reportedly ameliorates cerebral infarction in rats.16 Considering that administration of Ve analog SMTP-7 to tMCAo mice showed weak peroxynitrite reactions (Figure 2), it is believed that anti-inflammatory effects of SMTP-7 in the mice are at least in part caused by its direct antioxidative actions.
In the current in vitro study, we mainly used rt-PA at a concentration of 20 μg/mL, based on previous report that plasma concentrations of t-PA in an animal receiving a therapeutic dose of t-PA is capable of reaching such levels.17 In contrast to tPA, our previous in vitro study suggested that an effective concentration of SMTP-7 for fibrin degradation is approximately 100 μg/mL9; thus, it has been expected that SMTP-7 is functional at relatively high concentrations. Contrary to our expectation, these data indicated that SMTP-7 inhibits leukocyte/EC adhesion with an IC50 of 6.1 μg/mL (Figure 4); thus, the concentration that exerts an anti-inflammatory effect is much lower than that which exerts a thrombolytic effect. Moreover, our in vivo data suggested that administration of SMTP-7 at 10 mg/kg resulted in thrombolytic reperfusion (Figure 1). It is noteworthy that in vivo, an effective dose of rt-PA to lyse a clot is reportedly 10 mg/kg,18 indicating an equivalent thrombolytic effect of SMTP-7 and rt-PA in the in vivo conditions. Therefore, SMTP-7 is thought to be a unique thrombolytic agent that possesses robust thrombolytic and antioxidative activity.
These data suggested that administration of rt-PA to tMCAo mice led to increasing leukocyte/EC interactions, upregulation of inflammatory genes, and acceleration of cerebral infarction (Figure 2, 3). It has been documented that t-PA promotes MMP-9 secretion, NF-κB activation, apoptosis, and barrier dysfunctions in H/R-stimulated ECs.4,19 These results indicated that rt-PA facilitates cytokine-induced ROS generation (Figure 6) and H/R-induced VCAM-1 induction in ECs (Figure 4); thus, it is thought that rt-PA directly enhances proinflammatory signaling in ECs in the tMCAo mice. We believe that rt-PA is capable of accelerating cerebral infarction through proinflammatory responses independently of blood flow recovery when the ischemic region is highly restricted; however, reasons why rt-PA worsened cerebral infarction in the mice are currently obscure.
In conclusion, these data clearly indicate distinct effects of t-PA and SMTP-7 on cerebrovascular inflammation. It is likely that cytokine-mediated proinflammation in tMCAo mice is neutralized by the antioxidative thrombolytic agent SMTP-7, but not by rt-PA. Considering that SMTP-7 did not worsen tMCAo-induced cerebral infarction (Figure 2), we believe that SMTP-7 is a possible candidate drug for thrombolytic therapy. Future study will be needed to identify the cytoprotective effects of SMTP-7 on cerebral hemorrhage.
Sources of Funding
This study was supported in part by Showa University Special Grant-in-Aid for Innovative Collaborative Research Project [to H.O.], KAKENHI from Japan Society for the Promotion of Science (#20790178 to T.M., #21590240 to H.O.); the Japan Health Sciences Foundation (to H.O.); and “High-Tech Research Center” Project for Showa Universities from Ministry of Education, Culture, Sports, Science and Technology, 2000–2003 (to K. Ho).
The online-only Data Supplement is available at http://stroke.ahajournals.org/cgi/content/full/STROKEAHA.110.598359/DC1.
- Received July 29, 2010.
- Accepted November 9, 2010.
- © 2011 American Heart Association, Inc.
- Zhang RL,
- Zhang ZG,
- Chopp M,
- Zivin JA
- Miyazaki T,
- Honda K,
- Ohata H
- Miyazaki T,
- Honda K,
- Ohata H
- Tejima E,
- Katayama Y,
- Suzuki Y,
- Kano T,
- Lo EH